Historically humans have had to make decisions with too little data and guess. Now we have too much data and still end up guessing. How do you know when you have enough information to move on and stop analyzing? This was an issue that came up just in the last couple weeks as we went over a client's survey data. We are always looking for that next insight, that next pearl of wisdom that can be the "AH HA" moment of how to accomplish a goal. The problem is if our "AH HA" moment doesn't meet the climactic crescendo we're hoping for it feels we haven't found it yet and keep looking. Sound familiar?
Analysis paralysis is a common dilemma faced by many individuals, characterized by over-thinking and over-analyzing the data to the point where decision-making becomes exceedingly difficult. So how do we break this chain?
Strategies to overcome analysis paralysis
Start with questions.
What question are you trying to answer with your data? Create a list of 5-7 questions where the answer feels like your "AH HA" moment. Focus all analysis efforts on answering those questions first.
Keep them simple for starters:
- Why did the customer choose us?
- How did the customer find/discover us?
- Did the customer get what they needed and are they satisfied with their choice?
Ideally these questions are determined before you even write the first survey (if using survey data is the goal), so you can better craft your questions and understand the data more clearly afterwards.
If your data set is not survey data you should still start with your target questions first. After those are answered then you can look for deeper insights as time allows.
Step 1 of getting over Analysis Paralysis is to give yourself a basket of time to do your research. Once that time is up you stop--so long as you answered your priority questions. If not keep going. If you have time left you use that to look for other pearls of wisdom
Does an insight have an impact? That's a decision only you the analyst can answer. Choose yes or no and move on. If the answer is yes, write it down, and look for the next one. If the answer is no, make a note you looked into it and decided it wasn't relevant.
Be OK with the fact the answer may not exist in your data
It's common going through datasets when the answer just simply isn't there. It doesn't mean the insight doesn't exist, just that this particular dataset didn't capture it. The instinct many will face is to derive and assume an answer instead of saying "I don't know". If the question is unimportant then you can ignore and move on. If it turns out to be important then you can take steps to determine what data is needed to answer the question. Then you can figure out if that's data you have or data you need and continue accordingly.
Tell your narrative
Using your questions from earlier plan out your story so you can take your answers and plug them in the right place. This will also help you plan your time to ensure you leave enough time for writing the narrative and recommendations. If you couldn't directly answer one of your questions you can use the recommendation section to advise on new data to collect to answer the hanging question.
Analysis Paralysis hits most people at some point. If you reached a wall and can't continue further, feel free to schedule a consult and we may be able to help you find the narrative you're looking for, or take on the project for you.